Preprint · 2024 · Open access
Vision Transformers with Natural Language Semantics
arXiv (Cornell University) · Published 27 February 2024 · DOI 10.48550/arxiv.2402.17863
In brief
“Vision Transformers with Natural Language Semantics” is a 2024 preprint by Young Kyung Kim and 2 co-authors, published in arXiv (Cornell University). OpenAlex counts 12 citations as of 22 November 2026. It is open access: the full text is free to read at arXiv (Cornell University).
- Citations
- 12
- Most cited in
- 2024 (4)
- References
- 44
- Authors
- 3
Abstract
Tokens or patches within Vision Transformers (ViT) lack essential semantic information, unlike their counterparts in natural language processing (NLP). Typically, ViT tokens are associated with rectangular image patches that lack specific semantic context, making interpretation difficult and failing to effectively encapsulate information. We introduce a novel transformer model, Semantic Vision Transformers (sViT), which leverages recent progress on segmentation models to design novel tokenizer strategies. sViT effectively harnesses semantic information, creating an inductive bias reminiscent of convolutional neural networks while capturing global dependencies and contextual information within images that are characteristic of transformers. Through validation using real datasets, sViT demonstrates superiority over ViT, requiring less training data while maintaining similar or superior performance. Furthermore, sViT demonstrates significant superiority in out-of-distribution generalization and robustness to natural distribution shifts, attributed to its scale invariance semantic characteristic. Notably, the use of semantic tokens significantly enhances the model's interpretability. Lastly, the proposed paradigm facilitates the introduction of new and powerful augmentation techniques at the token (or segment) level, increasing training data diversity and generalization capabilities. Just as sentences are made of words, images are formed by semantic objects; our proposed methodology leverages recent progress in object segmentation and takes an important and natural step toward interpretable and robust vision transformers.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Kim, Y. K., Di Martino, J. M., & Sapiro, G. (2024). Vision Transformers with Natural Language Semantics. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2402.17863
MLA 9
Kim, Young Kyung, et al. “Vision Transformers with Natural Language Semantics.” arXiv (Cornell University), 2024. https://doi.org/10.48550/arxiv.2402.17863.
Chicago
Kim, Young Kyung, J. Matías Di Martino, and Guillermo Sapiro. 2024. “Vision Transformers with Natural Language Semantics.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2402.17863.
Harvard
Kim, Y. K., Di Martino, J. M. and Sapiro, G. (2024) ‘Vision Transformers with Natural Language Semantics’, arXiv (Cornell University). doi:10.48550/arxiv.2402.17863.
IEEE
Y. K. Kim, J. M. Di Martino, and G. Sapiro, “Vision Transformers with Natural Language Semantics,” arXiv (Cornell University), 2024, doi: 10.48550/arxiv.2402.17863.
BibTeX
@misc{kim2024vision,
title = {{Vision Transformers with Natural Language Semantics}},
author = {Kim, Young Kyung and Di Martino, J. Matías and Sapiro, Guillermo},
howpublished = {arXiv (Cornell University)},
year = {2024},
doi = {10.48550/arxiv.2402.17863},
url = {https://doi.org/10.48550/arxiv.2402.17863}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Citations by year
| Year | Citations | Share of the best year |
|---|---|---|
| 2023 | 1 | |
| 2024 | 4 | |
| 2025 | 4 | |
| 2026 | 3 |
Topic and keywords
Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- vision transformer
- semantic tokens
- image segmentation
- image tokenization
- inductive bias
- out-of-distribution generalization
- distribution shift robustness
- scale invariance
Work with this paper
Most-cited papers in Computer Science
- ImageNet classification with deep convolutional neural networksAlex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton · Communications of the ACM · 2017 · 114,726 citations
- Long Short-Term MemorySepp Hochreiter, Jürgen Schmidhuber · Neural Computation · 1997 · 109,097 citations
- A Mathematical Theory of CommunicationClaude E. Shannon · Bell System Technical Journal · 1948 · 91,736 citations
- Deep learningYann LeCun, Yoshua Bengio, Geoffrey E. Hinton · Nature · 2015 · 90,226 citations
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksShaoqing Ren, Kaiming He, Ross Girshick and 1 more · IEEE Transactions on Pattern Analysis and Machine Intelligence · 2016 · 74,143 citations
Questions about this paper
How do I cite this paper in APA style?
Kim, Y. K., Di Martino, J. M., & Sapiro, G. (2024). Vision Transformers with Natural Language Semantics. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2402.17863
How many times has this paper been cited?
12 times, according to OpenAlex on 22 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.
Is this paper free to read?
Yes. It is open access at arXiv (Cornell University): sign in and open the PDF from this page.
Who wrote this paper?
Young Kyung Kim, J. Matías Di Martino, Guillermo Sapiro.
What is the DOI of this paper?
10.48550/arxiv.2402.17863, permanently at https://doi.org/10.48550/arxiv.2402.17863. Sign in and open it from this page.
Bibliographic data from OpenAlex (CC0), last checked 22 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.